1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
|
#' Calculate the AIC for a column of an mmkin object
#'
#' Provides a convenient way to compare different kinetic models fitted to the
#' same dataset.
#'
#' @importFrom stats AIC BIC
#' @param object An object of class \code{\link{mmkin}}, containing only one
#' column.
#' @param \dots For compatibility with the generic method
#' @param k As in the generic method
#' @return As in the generic method (a numeric value for single fits, or a
#' dataframe if there are several fits in the column).
#' @author Johannes Ranke
#' @examples
#'
#' \dontrun{ # skip, as it takes > 10 s on winbuilder
#' f <- mmkin(c("SFO", "FOMC", "DFOP"),
#' list("FOCUS A" = FOCUS_2006_A,
#' "FOCUS C" = FOCUS_2006_C), cores = 1, quiet = TRUE)
#' # We get a warning because the FOMC model does not converge for the
#' # FOCUS A dataset, as it is well described by SFO
#'
#' AIC(f["SFO", "FOCUS A"]) # We get a single number for a single fit
#' AIC(f[["SFO", "FOCUS A"]]) # or when extracting an mkinfit object
#'
#' # For FOCUS A, the models fit almost equally well, so the higher the number
#' # of parameters, the higher (worse) the AIC
#' AIC(f[, "FOCUS A"])
#' AIC(f[, "FOCUS A"], k = 0) # If we do not penalize additional parameters, we get nearly the same
#' BIC(f[, "FOCUS A"]) # Comparing the BIC gives a very similar picture
#'
#' # For FOCUS C, the more complex models fit better
#' AIC(f[, "FOCUS C"])
#' BIC(f[, "FOCUS C"])
#' }
#'
#' @export
AIC.mmkin <- function(object, ..., k = 2)
{
# We can only handle a single column
if (ncol(object) != 1) stop("Please provide a single column object")
n.fits <- length(object)
model_names <- rownames(object)
code <- paste0("AIC(",
paste0("object[[", 1:n.fits, "]]", collapse = ", "),
", k = k)")
res <- eval(parse(text = code))
if (n.fits > 1) rownames(res) <- model_names
return(res)
}
#' @rdname AIC.mmkin
#' @export
BIC.mmkin <- function(object, ...)
{
# We can only handle a single column
if (ncol(object) != 1) stop("Please provide a single column object")
n.fits <- length(object)
model_names <- rownames(object)
code <- paste0("BIC(",
paste0("object[[", 1:n.fits, "]]", collapse = ", "),
")")
res <- eval(parse(text = code))
if (n.fits > 1) rownames(res) <- model_names
return(res)
}
|
